Agar Plate Dilution Method for Routine Antibiotic
Susceptibility Testing in a Hospital Laboratory
KENNETH C. HALTALIN, M.D.,
ANNE H. MARKLEY, B.S., B.A.,
MT(ASCP),
AND EDYTHE WOODMAN
The Microbiology Laboratory, Children's Medical Center of Dallas, and Department
of Pediatrics, the University of Texas Southwestern Medical School
at Dallas, Dallas, Texas
ABSTRACT
Haltalin, Kenneth C , Markley, Anne H., and Woodman, Edythe: Agar plate
dilution method for routine antibiotic susceptibility testing in a hospital
laboratory. Am. J. Clin. Pathol. 60: 384-394, 1973. The agar plate dilution
method for routine antibiotic susceptibility testing has been used in this hospital laboratory since 1968. T h e experience has met with the approval of
the laboratory staff, and the technic has provided clinicians with more precise
and useful information than was formerly obtained with disk sensitivity methods. Two studies were performed to document the reliability of our method.
Testing of 60 organisms by both agar dilution and broth dilution methods
showed that inhibitory concentrations were identical or varied by only one
dilution, with a single exception. T o compare results obtained using antibiotic laboratory standards and commercial intravenous preparations, 120
strains were tested in duplicate. Minimal inhibitory concentrations with
both forms of drugs were similar or varied by only one dilution in all cases.
A "time-work" study showed that the agar dilution method can be performed
economically in a routine laboratory setting. The agar dilution method is
appropriate for use in hospital laboratories and has distinct advantages over
the disk agar diffusion technic.
T H E VARIOUS TECHNICS available for antimicrobial susceptibility testing in diagnostic microbiology laboratories have been
critically evaluated in two recent monographs.3- 5 Virtually all clinical laboratories
use a disk agar diffusion method such as
the one described by Bauer and associates.2
Determination of antimicrobial susceptibility by dilution technics is assumed by
most technologists and laboratory directors
to be too complicated, expensive, and time-
Received September 12, 1972; received revised
manuscript October 12, 1972; accepted for publication November 15, 1972.
Address reprint requests to: Kenneth C. Haltalin,
M.D., Department of Pediatrics, The University of
Texas Southwestern Medical School at Dallas, 5323
Harry Hines Boulevard, Dallas, Texas 75235.
consuming for routine use in clinical laboratories. Several years ago, the agar plate
dilution method was semiautomated by the
introduction of a simple, inocula-replicating apparatus by Steers and associates,8
thereby making the technic more feasible
for use in diagnostic laboratories. In spite
of this, the routine use of the agar dilution
method has apparently remained confined
to very few clinical laboratories 6 and to
research units.
In mid-1968 an agar plate dilution
method utilizing the device of Steers and
associates replaced the disk technic as the
principal method of performing susceptibility testing in the microbiology labora-
384
September 1973
ANTIBIOTIC SUSCEPTIBILITY TESTING
tory of Children's Medical Center of Dallas.
This communication describes our procedure, evaluates its reliability and efficiency,
and summarizes selected results.
Materials and Methods
Hospital and Microbiology
Laboratory
Children's Medical Center is a 122-bed
hospital with full medical and surgical
services and is a major teaching institution
for house staff and medical students of the
Departments of Pediatrics and Surgery of
the University of Texas Southwestern Medical School at Dallas. During 1971 there
were 6,624 hospital admissions (60% private and 40% staff) and 63,976 outpatient
visits. There are 176 surgeons and 169 pediatricians on the attending staff.
The microbiology laboratory is staffed by
a supervisor, two medical technologists, one
clerk-laboratory assistant, and a part-time
media-maker. The laboratory is directed by
the chief of pathology; an advisory role is
assumed by a member of the pediatric infectious disease service. Two technologistsin-training are assigned to the laboratory
at all times and a resident physician or a
medical student is present 1 day a week for
instruction. Approximately 60 specimens
are received each day and about 20 organisms are tested for antibiotic susceptibility daily Monday through Friday.
3S5
species, and unusual organisms are tested
with both sets. The commercial preparations of antibiotics for intravenous administration used in the tests are obtained from
the hospital pharmacy at a cost of approximately $20.00 per week. Chloramphenicol
is supplied gratis as the free salt by ParkeDavis. (Chloramphenicol succinate, the
commercial preparation for intravenous administration, has no antibacterial action
in vitro).
Once a week fresh antibiotic solutions
are prepared. For chloramphenicol, 100 mg.
of drug are dissolved in 2 ml. of 95% ethyl
alcohol. Other antibiotics are reconstituted
with sterile distilled water. Serial dilutions
are made with distilled water to give the
desired concentrations. One milliliter of a
given concentration of antibiotic is added to
99 ml. of sterile Oxoid Sensitivity Test
Medium* contained in an 8-ounce media
bottle held in a water bath at 45 to 48 C.
Twenty milliliters of agar are poured into
each of five petri plates (100 mm. X 15
mm.). Final antibiotic concentrations in
agar, except for gentamicin and penicillin,
are 20, 10, 5, 1.25, and 0.3 fig. per ml. Agar
concentrations of gentamicin are 10, 5, 2.5,
1.25, and 0.3 /xg. per ml., and those of penicillin are 2.5, 1.25, 0.6, 0.2, and 0.08 units
per ml. Plates are left at room temperature
overnight to dry the agar surfaces and then
stored at 4 C. All plates are used within 7
days. Double sets of gentamicin and cephAgar Dilution Method
alothin
plates are prepared each week, since
Gram-negative organisms are tested with
these
drugs
are used in both Gram-negaampicillin (Ayerst Laboratories), cephalotive
and
Gram-positive
sets. Growth conthin (Lilly & Company), chloramphenicol
trol
plates
are
poured
without
antibiotics.
(Parke-Davis & Company), kanamycin (BrisIn the morning, seed cultures are pretol Laboratories), polymyxin B (Pfizer Laboratories), tetracycline (Lederle Laborato- pared by inoculating organisms selected for
ries), and gentamicin (Schering Corpora- testing into screw-top tubes containing 5
tion) ("Gram Negatve Set"). Gram-positive ml. of Oxoid Sensitivity Test Broth and inorganisms are tested with penicillin G cubated at 37 C. for 4 hours. After incuba(Pfizer Laboratories), methicillin (Bristol tion 0.01 ml. of broth is removed with a
Laboratories), erythromycin (Abbott Lab- calibrated loop and placed into 1 ml. of
oratories), cephalothin and gentamicin
• Colab Laboratories, Inc., 3 Science Road, Glcn("Gram Positive Set"). Enterococci, Listeria wood Science Park, Glenwood, Illinois 60425.
386
A.J.CP.—Vol.
HALTALIN ET AL.
60
CHILDREN'S MEDICAL CENTER
ANTIMICROBIAL SUSCEPTIBILITY REPORT
CULTURE NO.:
NAME
DATE OF CULTURE:
RECORD NO.
SOURCE:
LOCATION^
'--
ORGANISM:_
DRUG
PLATE DILUTION SUSCEPTIBILITY
DISC
Erythromycin (mcg/ml)
N.I.
20
10
5
1.25
<0.3
Methicillin (mcg/ml)
N.I.
20
10
5
1.25
<0.3
Penicillin (Units/ml)
N.I.
2.S
1.2S
0.6
0.2
<0.08
Cephalolhin (mcg/ml)
N.I.
20
10
5
1.25
<0.3
Ampicillin (mcg/ml)
N.I.
20
10
5
1.25
<0.3
Chloramphenicol (mcg/ml)
N.I.
20
10
5
1.25
<0.3
Kanamycin (mcg/ml)
N.I.
20
10
5
1.25
<0.3
Polymyxin B (mcg/ml)
N.I.
20
10
5
1.25
<0.3
Tetracycline (mcg/ml)
N.I.
20
10
5
1.25
<0.3
Gentamicin (mcg/ml)
N.I.
10
5
2.5
1.25
<0.3
I
s
s
s
s
s
s
s
s
s
I R
I R
1 R
I R
I R
I R
1 R
1 R
I R
1
II
nesistant
Range
S 1 R
FIG. 1. Facsimile of form used to
report antimicrobial susceptibility
results. "N.I." stands for no inhibition at the highest concentration
of drug tested. MIC's are indicated
by circling the appropriate value.
Susceptibility with disc agar diffusion testing is reported as either
susceptible (S), intermediate (I), or
resistant (R).
Susceptib le
Range
sterile phosphate-buffered saline solution
(pH 7.4). Seed cultures of Pseudomonas species are not diluted.
Plates are inoculated with an inoculareplicating apparatus, 8 thereby allowing the
simultaneous inoculation of 32 organisms
onto each plate. Thus, a total of 64 organisms can be tested each day, since Gramnegative and Gram-positive sets are inoculated separately. Inoculating wells are filled
with the diluted seed cultures with sterile
Pasteur pipettes. Each inoculating prong
delivers approximately 0.01 ml., containing
10* to 103 viable organisms. After inoculation, plates are left in the upright position
for 30 min. to insure complete absorption
of the inocula and then inverted and incubated at 37 C. for 16 hr. The lowest concentration of a given drug resulting in complete inhibition of growth, a barely visible
haze of apparent growth, or a single pinpoint colony is reported as the minimal inhibitory concentration (MIC). 1
A facsimile of the antibiotic susceptibility report form is shown in Figure 1. Or-
ganisms with MIC's of < 10 jug. per ml.
with cephalothin, kanamycin, ampicillin,
methicillin, and chloramphenicol are defined as susceptible. For gentamicin, polymyxin B, tetracycline and erythromycin,
the level of susceptibility is set one dilution lower, at < 5 /*g. per ml. A MIC of
< 0.6 units per ml. indicates susceptibility
to penicillin.
For quality control, two stock strains of
coagulase-positive staphylococci, one strain
of E. coli, and one strain of Klebsiella pneumoniae with known, diverse susceptibility
patterns are tested each day. I n addition,
viable colony counts are performed periodically on diluted seed cultures selected at
random to insure that inocula contain the
desired number of organisms.
The disk diffusion technic of Bauer and
co-workers2 is used in place of the agar dilution method in the following circumstances: (1) routinely to test susceptibility
of enteropathogenic E. coli to neomycin,
(2) routinely to test susceptibilities of Proteus and Pseudomonas species to carbeni-
September
1973
cillin, (3) occasionally to test susceptibilities of an isolate available for testing on
Saturdays, and (4) occasionally to test susceptibilities of an isolate to infrequently
used drugs not included in the test battery.
Evaluation
387
ANTIBIOTIC SUSCEPTIBILITY TESTING
Studies
For the purposes of this report three
studies were performed:
(1) Sixty bacteria chosen at random were
tested by the broth dilution technic and
the results correlated with those previously
obtained by routine agar dilution testing.
Thirty-five Gram-negative organisms comprising five strains of each of the following
seven types of bacteria were tested: Klebsiella, Enterobacter, E. coli, Ps. aeruginosa,
Shigella, Salmonella and Proteus. Twentyfive isolates of coagulase-positive staphylococci were tested. Broth dilution testing
was done in the pediatric infectious disease
research laboratories. Antibiotics were diluted serially in screw-top tubes containing
5 ml. amounts of Oxoid Sensitivity Test
Broth to obtain the same concentrations
present in agar. Initial seed cultures were
prepared as they were for the agar dilution
method. Diluted seed cultures were made
by inoculating 0.01 ml. of the initial seed
cultures into 10-ml. portions of sterile phosphate-buffered saline solution (pH 7.4); 0.1
ml. amounts of these cultures were then
transferred to the antibiotic-containing
broths by sterile pipettes. Viable colony
counts were determined for all diluted seed
cultures tested by both agar and broth technics to insure that inoculum sizes were comparable. Tubes were incubated at 37 C. for
16 hr. The lowest concentration of antibiotic resulting in an absence of visible
growth was recorded as the MIC.
(2) T o document that results obtained
with commercial antibiotic preparations
compare favorably with those obtained
using antibiotic laboratory standards, 120
organisms were tested with plates prepared
with both forms of drug. Polymyxin B was
not evaluated because a laboratory standard
was not available, nor was chloramphenicol
tested in duplicate since the laboratory
standard is used for routine testing.
(3) A "time-work" study was carried out
to determine the amount of time the agar
dilution method demands of the laboratory
staff during an average week.
Dissemination of Susceptibility
tion to Physicians
Informa-
Copies of the antibiotic susceptibility report forms are sent to the infectious disease
consultant once a month. Results are tabulated according to sources of cultures and
organisms isolated. Every 6 months (January and August), cumulated summaries of
selected information with interpretative
comments are sent to the attending physicians and house staff.
Results
Comparison of Broth and Agar
Dilution Methods
Results obtained testing Gram-negative
organisms by tube and plate methods are
compared in Figure 2 (tetracycline, ampicillin, chloramphenicol, and cephalothin)
and Figure 3 (polymyxin B, gentamicin,
and kanamycin). Except for one isolate
tested with chloramphenicol, all organisms
had identical MIC's or varied by only one
dilution when tested by the two methods.
If it is assumed that the broth dilution
technic is the more accurate of the two
methods, the results can be evaluated in a
more clinically relevant manner by asking
the following question: how many organisms tested by the agar technic would
have been interpreted by the physician as
resistant when the broth method indicated
susceptibility, and vice versa? For ampicillin and gentamicin, such a discrepancy
would not have occurred; for tetracycline,
polymyxin B and kanamycin it would have
occurred with one isolate each; for cephalo-
388
E
5
A.J.CP.—Vol. 60
HALTALIN ET AL.
y
Tetracycline
>20
-
20
/
20
•
10
c
o
Chloramphenicol
>20
/
10
5
5
'•7-'
1.25
50.3
/
§
;
,
,
50.3 1.25
5
1.25
50.3
10
i
i
20
>20
50.3
1.25
5
10
>20
20
'c
O
O
X
tLU
5
>20
•
/
5
^
/
*
1.25
/
V
/
20
• y
10
o
10
5
/
: _z
:0.3
50.3
/ . ,
<
50.3
.
1
1
1
5
10
20
»20
1.25
/
i
50.3
.
v
/
y
1.25
O
UJ
/
>20
20
Z
_J
Cephalothin
Ampicillin
2
Fie. 2. Correlation of
results of agar plate dilution and broth dilution
tube methods obtained
testing 35 Gram-negative
organisms with four antibiotics.
i
i
i
1.25
5
10
i
>20
20
TUBE DILUTION METHOD (Minimal Inhibitory Concentration-/to/ml)
Polymyxin B
4.
i
c
o
/
•20
«o
?0
c
10
w
•
u
c
o
Inh ito
>•
w
j
'
//
5
1 'i"l -
50,3
/
y
/'
i
i
i
i
i
1.25
5
10
20
»20
ui;
Gentamicin
>I0
-
>?0
10
-
20
ION
^
1.25
FIG. 3. Correlation of results of agar plate dilution and broth dilution
tube methods obtained
testing 35 Gram-negative
organisms with three antibiotics.
m
w
H8
Kanamycin
/
/
/
5
2.5
kTE DIL
i
50.3
S
2
J
/
/
- A n
r
o
w
Q
O
X
1Ul
.m i
/
S
10
-
5
wot
l?5
50.3
-
/
50.3
/
,
,
,
'
i
50.3
1.25 2.5
5
10
»I0
r
t
50.3
\
1.25
i
5
i
i
i
10
20
>20
TUBE DILUTION METHOD (Minimal Inhibitory Concentration-^ig/ml)
September 1973
389
ANTIBIOTIC SUSCEPTIBILITY TESTING
v.
Penicillin
/
>2.5
-
/
2.5
/
.
.
/
0.2
Fic. 4. Correlation of results of agar plate dilution and broth dilution
tube methods obtained
testing 25 coagulase-positive staphylococci with
four antibiotics.
/ .
/
0.08
/
r 50.08 0.2 0.6
1.25 2.5 >2.5
Methicillin
a
o
I
/
>20
/
20
/
10
o
5
'
I-
1.25
Q
UJ
I-
<
0.
/
** tBB
50.3
/
1
<0.3
1
1
1
1
1.25
5
10
20
p
>20
<0.3
1.25
5
10
20
>20
TUBE DILUTION METHOD (Minimal Inhibitory Concentration-/ig/ml)
thin with two isolates; for chloramphenicol
with four organisms. It other words, of 135
tests (35 organisms tested with seven drugs)
6.6% (nine tests) were interpreted as resistant by the agar method when they
would have been considered susceptible by
the broth technic. In only one case (0.7%),
with tetracycline, was an organism assumed
to be susceptible by agar testing when it
was resistant by the broth method.
Results obtained testing coagulase-positive staphylococci by the two methods with
penicillin, methicillin, erythromycin, and
cephalothin are compared in Figure 4. All
strains had identical MIC's or varied by
only one dilution when tested by the two
methods. Correlation with methicillin and
cephalothin was exceedingly accurate, with
88% (22/25) of methicillin tests and 92%
(23/25) of cephalothin tests having identical MIC's by the two methods. Of 100
tests (25 organisms tested with four drugs),
only 2%, representing two erythromycin
results, would have been interpreted by the
clinician as resistant when they were susceptible by the broth method. Conversely,
only 3 % of tests (all with penicillin) would
have been interpreted as susceptible by
agar testing when they were resistant by
the broth method. Results of gentamicin
testing (not depicted graphically) showed
that 80% (20/25) of isolates had identical MIC's by the two methods and 20%
(5/25) varied by one dilution. In no case
would the interpretation of resistance or
susceptibility have been at variance by
either method.
Comparison of Antibiotic Laboratory
Standards and Commercial
Antibiotics
Table 1 shows the correlation of the
MIC's obtained using both forms of drug.
Without exception, all MIC's were identical or varied by only one dilution. Identical
results were seen in 90% or more of tests
390
HALTALIN ET AL.
A.J.C.P.—Vol.
60
Table 1. Correlation of Minimal Inhibitory Concentrations Obtained with Antibiotic
Laboratory Standards and with Commercial Antibiotic Preparations
for Intravenous Administration
Minimal Inhibitory Concentrations
Number and Type of
Organisms Tested
Antibiotic
70 gram-neg. rods
39 gram-pos. cocci
70 gram-neg. rods
39 gram-pos. cocci
70 gram-neg. rods
70 gram-neg. rods
70 gram-neg. rods
50 gram-pos. cocci
50 gram-pos. cocci
50 gram-pos. cocci
Gentamicin
Cephalothin
Ampicillin
Kanamycin
Tetracycline
Erythromycin
Methicillin
Penicillin
with cephalothin, kanamycin and tetracycline; in 7 1 % and 78%, respectively, with
ampicillin and erythromycin; and in between 50 and 60% of tests with gentamicin,
methicillin, and penicillin. With two antibiotics, gentamicin and ampicillin, the
one-dilution variation was almost exclusively in the same direction, with the laboratory standards producing MIC's one dilution lower than the commercial preparations. For gentamicin the figures were 45
of the 46 variations, and for ampicillin they
were 18 of 20 variations.
Time-Work
Evaluation
The time required by the laboratory
staff to perform the procedures connected
with the agar dilution method for a typical
week is shown in Table 2. T h e features
unique to the method, dilution of antibiotics and use of the replicator, occupy only
one-fifth of the total time expended. The
other items of work, with appropriate modifications, are necessary for the performance of any testing method.
Six-month
Summaries
Summaries of results for the 6-month period from July 1 to December 31, 1971, that
Onedilution
Identical Variation
(%)
(%)
58
42
94
6
71
90
91
78
54
56
29
10
9
22
46
44
were circulated to the house staff and attending physicians are presented in Tables
3 and 4.
Discussion
T h e agar dilution method using antibiotic
preparations designed for intravenous administraiton is a reliable, efficient, and financially feasible way to perform routine
antibiotic susceptibility testing in our hospital laboratory. Results compare favorably
with those obtained with broth dilution
testing and with agar dilution testing using
antibiotic laboratory standards. Intravenous antibiotic preparations obtained from
the hospital pharmacy are recommended
for routine agar dilution testing,1 but variations in potency may ocurr from one manufacturer to another because of such factors
as "overfilling" of vials. Even though it is
unlikely that such variations would affect
results significantly, it would be ideal if
standard laboratory preparations were commercially available from suppliers such as
those who manufacture disks. At the present time laboratory standards must be obtained directly from each pharmaceutical
house. This would be a major step toward
developing the agar dilution method into
September 1973
391
ANTIBIOTIC SUSCEPTIBILITY TESTING
a fully standardized technic suitable for
widespread use.
The agar dilution method has been accepted enthusiastically by our laboratory
personnel, who believe, on the basis of past
experience, that agar dilution testing is no
more, and probably less, time-consuming
to perform than the disk agar diffusion
technic of Bauer and associates.3 Although
we did not evaluate the differences between
the agar dilution method and this disk
technic, we feel that it is appropriate to
contrast some of the features of each, since
it is used so extensively. In our opinion,
the agar dilution method possesses several
advantages. T o begin with, our dilution
method utilizes Oxoid Sensitivity Test Medium instead of Mueller-Hinton medium.
The Oxoid preparation is a clearer medium that generally yields better bacterial
growth, supports optimal growth of streptococci without additives, and decreases somewhat the swarming character of Proteus
species.* Results of disk testing with the
Oxoid medium have been reported to be
comparable to those obtained with MuellerHinton medium* but it is not recommended for routine use because the disk
technic has been standardized with the latter medium.
The depth of the agar for the disk technic is important, and should be 5 to 6 mm.2
Variations in depth may give rise to misleading results, since antibiotics diffuse
downward as well as outward. Depth of the
agar is not a critical factor with the agar
dilution method because the antibiotics are
evenly concentrated throughout the medium. According to Bauer and colleagues,2
poured plates for the disk technic should
be used within 4 days after preparation,
while plates containing antibiotics incorporated in agar may be stored longer. The
antibiotics used in our test battery, with
the exception of gentamicin, have been
evaluated by others with respect to their
stability in agar after refrigerated storage
Table 2. Amount of Time Required per Week
to Perform Agar Plate Dilution
Testing*
Procedure
Media and antibiotics
Preparation of media
Dilution of antibiotics
Pouring and labeling of plates
Preparation of seed cultures
Performance of tests
Loading of inoculating wells
and replication
Reading of results
Recording and reporting of results
TOTAL
Time
(Hr.)
1
1$
2
3J
1J
1}
3J
14
* Based on an average of 100 organisms per week.
for various periods of time.7 Significant loss
of activity occurred with none after 1 week
and only with penicillin and ampicillin
after 4 weeks of storage. Our experience
with gentamicin shows that it is stable in
agar for at least 10 days.
The inoculum delivered to the agar surface in the disk technic cannot be quantitated as accurately as it can in the agar
dilution method because the amounts of
seed culture adhering to the applicator
swab vary from one culture to another.
In our opinion the procedures connected
with the disks themselves are a vexing part
of the disk technic. They must be stored
properly to avoid loss of potency, expiration dates must be checked, and they must
be placed carefully in appropriate positions
on the agar surface. Furthermore, we believe that with agar dilution testing reading
results is simpler and more accurate than
with the disk technic. With agar dilution
testing results are read as either "growth"
or "no growth" for a given concentration
of antibiotic. Occasionally a barely visible
haze or a single pinpoint colony may appear at the site of inoculation; such occurrences are interpreted as no growth. 8 With
the disk technic, precise measurements of
zone diameters must be made, which at
98
99
81
16
95
95
5
131
56
123
57
191
Klebsiella
Enterobacter
Shigella
Salmonella
Pseudomonas aeruginosa
100
98
98
93
98
20
67
15
64
95
99
96
Proteus
Indol-positive
Indol-negative
81
83
87
8
93
84
89
86
80
92
87
92
82
Kanamycin
<10/ig./ml.
83
64
89
82
10
27
97
86
8
86
63
99
100
90
32
27
5
74
69
72
Tetracycline
<5/«g./ml.
33
84
65
75
81
Ampicillin
<10 M g./rnl.
0
0
92
98
96
Polymyxin B
<5/xg./ml.
Percentage Susceptible to Concentrations of Antibiotic Indicated
Gentamicin
< 5 ng./w\.
78
167
67
Cephalothin
<10Mg./ml.
Escherichia coll
Enteropathogenic
Urine
Other sources
Type of Bacteria
Number
Tested
Table 3. Plate Dilution Antibiotic Susceptibilities of Gram-negative Bacteria Isolated at
Children's Medical Center from July 1 to December 31, 1971
9
100
97
83
94
60
88
92
94
94
Chloramphenicol
< 10 iig./ml
393
i
o
r
ft
8S
s?
o «
QJ
a .a .o
CJ
Not
tested
o
u
o
1
o
-H
/ H
bo
a.
<u
CO
OJ
4-)
c
emoly tic
CO
(J
a
^H
.y
*-3
o
:§
PQ - i
c
JS
aj
"2 -"
CU
<
tn
o
O
c
*-*
<u
u
isl
4 *
88
l-H T ^
c
'p
S"8
O -4-*
<^l QJ
«°
*
Pi
o >o
"^
"o
cocc
emb
•r-
I-I
ei .S O "H
W E
V I
o is
/<
»
0>
4-)
u
CO ^H
c
o
(J
o
•4-1
J? >•
.£ 3
'8 T
k ~o
ft
CJ
a
o
tn
0)
u _
CJ
P-i
cu
<u
:~s
a "
H
CJ T3
—i
If)
•£
O
Ov Os
ON C \ C\
O
ro
O \ 0 \ 0
O
PO
• ^ CN
S'SVI>
,
o
_:
C4 C ° E
3
4-*
o
.y
en " O
i .s .„ i
J2
V3
v
~X
3
*->
60 C
a) i >
CJ
^
*
^
to
C
CJ
,
2 "cj
= to -5
ja
c
M O O \
Os Ov Os
8S
'O
Os Os Q
Os Os O
O
r o 0 0 PO
r - r*- 0 0
Q PO
O fO
a
w
-
c
|2"S
§
PO
*
c
3 -^
CO U
a *>. "H
•3 oj
O -o
u o ;
'gvip
PM
^
13 «J
'3
c4
<
«
r—i
C
O
u "O
CJ
J D CJ
^
-*->
3
r3
^
25
^
QJ
'D'
•^
•a
O
cu
Pn
CN
>
p
'lat
times may be difficult because of hazy zone
perimeters or asymmetrical zone areas even
with use of an electronic zone reader. Measurements must then be checked against
standards for each antibiotic to determine
whether the organism is susceptible, of intermediate susceptibility, or resistant.
Last, we believe that the most important
advantage of the agar dilution method is
that results are reported as MIC's of a
given antibiotic. Thus, clinicians are able
to correlate these values with serum levels
of antibiotic achievable in patients following various doses given by the various
routes of administration.
T o avoid problems of acceptance and
understanding by physicians, a written
explanation outlining the agar dilution
method was sent to both the house staff and
the attending staff before institution of the
technic. No complaints concerning either
the type and number of antibiotics tested
or the manner in which results are reported
have been received. T h e house staff is carefully instructed regarding the interpretation of minimal inhibitory concentrations
in the light of attainable levels of antibiotic in serum and other fluids and the
individual patient's disease process. If a
member of the attending staff is uncertain
how to interpret a result, he usually will
consult a house officer or check with a member of the infectious disease service.
The types of antibiotics that are prescribed in our hospital have been influenced by the agar dilution method. We
have found that physicians are unlikely to
use drugs that are not included in the
agar dilution test battery, with the exception of such drugs as neomycin for diarrheal disease due to enteropathogenic E.
coli and carbenicillin for infections caused
by Ps. aeruginosa and certain Proteus species; disk testing is performed routinely
under these circumstances. Physicians may
request that disk testing be performed with
other agents, but rarely do so.
Chloram
phenico
<10
ANTIBIOTIC SUSCEPTIBILITY TESTING
Table
September 1973
•c
n
a
"o
a)
E
£Q
2 S
5O
rt
CJ
Q.
>>
H
_ a.
O -12 CJ J3
O
«-<
CJ flJ
o
^?
gO g.
Z.Z S
394
HALTALIN ET AL.
It could be argued that other drugs
should be added to the test battery and
that some currently included should be deleted. We believe that the drugs tested
should be appropriate for the particular
population of a given hospital. Furthermore, the total number of drugs tested
and the antibiotic concentrations evaluated
should be realistic so far as the laboratory
is concerned. At this point, the drugs included in our battery and the concentrations used are appropriate for our hospital.
However, the composition of the battery should be reviewed periodically and
changes made when indicated.
The total number of organisms that can
be tested each day, including stock control
strains, without increasing the amount or
cost of materials is 64; 32 in the Gramnegative set and 32 in the Gram-positive
set. This potential is sufficient to take care
of our needs for the foreseeable future.
When the stage is reached where 32 Gramnegative organisms are tested each day, "restraining rings"! will have to be applied
to the agar surrounding Proteus inocula to
inhibit swarming completely. At the present time Proteus isolates are segregated to
one portion of the plate and spaced appropriately.
Adoption of the agar dilution method for
use in hospitals with larger bacteriology
services than ours is feasible.6 For example,
the laboratory of Parkland Memorial Hospital in Dallas has used an agar dilution
f Available as "Raschig rings" from Scientific Apparatus Co., Bloomfield, N. J.
A.J.C.P.—Vol.
60
method for several years and performs an
average of 60 susceptibility tests daily.
Until recently plates in that laboratory
were inoculated using a swab technique; inoculation is now performed with the replicating device of Steers and associates.8
Acknowledgments. Patricia A. Brooks, M.T.
(ASCP), Karen B. Mover, B.S., and Arthur G.
Weinberg, M.D., Director of Laboratories, cooperated in this study. Helen T. Kusmiesz, R.N., tabulated the susceptibility information that was sent
to the physicians.
References
1. Anderson TG: Testing of susceptibility to antimicrobial agents and assay of antimicrobial
agents in body fluids, Manual of Clinical Microbiology. Edited by JE Blair, EH Lennette
and JP Truant. First edition. Bethesda, Md.
American Society for Microbiology, 1970, pp
299-307
2. Bauer AW, Kirby WMM, Sherris JC, et al: Antibiotic susceptibility testing by a standardized
single disk method. Am J Clin Pathol 45:493496, 1966
3. Ericson HM, Sherris JC: Antibiotic sensitivity
testing: Report of an international collaborative study. Acta Pathol Microbiol Section B
suppl 217:1-90, 1971
4. Esser VM, ElefsOn DE: Experiences with the
Kirby-Bauer method of antibiotic susceptibility
testing. Am J Clin Pathol 54:193-198, 1970
5. Gavan TL, Cheatle EL, McFadden HW Jr: Antimicrobial Susceptibility Testing. Chicago, American Society of Clinical Pathologists Commission on Continuing Education, 1971, pp 1-243
6. Marsh HH III: Automation in antimicrobial susceptibility testing. Antimicrobial Susceptibility
Testing. Edited by TL Gavan, EL Cheatle HW
McFadden Jr. Chicago, American Society of
Clinical Pathologists Commission on Continuing Education, 1971, p 237
7. Ryan KJ Needham GM, Dunsmoor CL, et al:
Stability of antibiotics and chemothcrapeutics
in agar plates. Appl Microbiol 20:447-451, 1970
8. Steers E, Foltz EL, Graves BS: Inocula replicating apparatus for routine testing of bacterial
susceptibility to antibiotics. Antibiot Chemother
9:307-311, 1959